Changelog
Release notes for InferLlama. For the full commit history, see GitHub releases ↗.
Admin dashboard with user management and HuggingFace sync
Paginated admin user list with search
Change password and delete account from Settings
Storage quota usage display in Settings
API key hashing fixed (SHA-256 consistency)
Server-side logout properly blacklists tokens
MinIO bucket set to private; presigned URLs for downloads
Focus trap in mobile filter drawer
Respects prefers-reduced-motion system setting
P2P model distribution via WebTorrent
Playground with streaming, conversation history, and keyboard shortcuts
Model upload with GGUF metadata extraction
OpenAI-compatible chat completions API
User profiles with public model listings
Rate limiting per-user and per-IP
Model registry with search, filter, and sort
JWT authentication with 15-minute access tokens
Account lockout after 5 failed login attempts
Password reset via email
API key management in Settings
Initial open-source release
FastAPI backend with PostgreSQL and Redis
Cloudflare R2 / MinIO object storage integration
Next.js 14 web application
Want to see what's coming next?
View roadmap on GitHub